An incidence survey of childhood rhabdomyosarcoma (RMS) has been conducted covering the State of North Carolina during the period 1967-1976. This was initiated by the clinical observation of a possible "epidemic" of RMS in the state. Preliminary data suggest that rather than an epidemic, there has been a "straight-line," almost 4-fold increase in RMS incidence over the 10-year study period. The incidence rate reached is greater than previously reported estimates based on mortality statistics but the same as estimated incidence data from the Third National Cancer Survey. It also appears that males are affected more than females (M/F equals 1.7) and blacks more than whites (B/W equals 1.3). In previous studies the excess of males was less marked (M/F equals 1.0-1.4) and whites experienced higher risks than blacks. A case-control approach is currently being used to study the "descriptive epidemiology" of RMS as well as associations between RMS and potential antenatal, childhood, familial, dietary and environmental etiologic exposures.